JP2007520948A - Image encoding method, image decoding method, image encoding device, image decoding device, and program - Google Patents

Abstract

  The image encoding method according to the present invention encodes a quantization matrix for encoding an image by orthogonal transform and quantization in block units and deriving a quantization step corresponding to each frequency of the orthogonal transform coefficient. In the encoding method, for each frequency component of the quantization matrix, a difference value between a frequency component and a predetermined value corresponding to the frequency component is obtained, the difference value is encoded into a variable length code, and the variable length code Has a code length that is shorter or the same as the difference value is smaller.

Description

  The present invention relates to an image encoding method for efficiently compressing a moving image, an image decoding method for correctly decoding the image, an image encoding device, an image decoding device, and a program.

  In recent years, with the era of multimedia that handles voice, images, and other pixel values in an integrated manner, conventional information media, that is, means for transmitting information such as newspapers, magazines, televisions, radios, and telephones to humans have been developed. It has come to be taken up as a target. In general, multimedia refers to not only characters but also figures, sounds, especially images, etc. that are associated with each other at the same time. It is an essential condition.

  However, when the amount of information possessed by each information medium is estimated as the amount of digital information, the amount of information per character in the case of characters is 1 to 2 bytes, whereas in the case of speech, 64 Kbits (phone quality) In addition, for a moving image, an information amount of 100 Mbits (current television reception quality) or more per second is required, and it is not realistic to handle the enormous amount of information in the digital format as it is with the information medium. For example, videophones have already been put into practical use by the Integrated Services Digital Network (ISDN) with a transmission speed of 64 Kbit / s to 1.5 Mbit / s, but the video from the TV camera is sent as it is via ISDN. It is impossible.

  Therefore, what is needed is information compression technology. For example, in the case of videophones, videos of H.261 and H.263 standards recommended by the ITU-T (International Telecommunication Union Telecommunication Standardization Sector) Compression techniques are used. Further, according to the MPEG-1 standard information compression technology, it is possible to put image information together with audio information on a normal music CD (compact disc).

  Here, MPEG (Moving Picture Experts Group) is an international standard of moving picture signal compression standardized by ISO / IEC (International Electrotechnical Commission International Electrotechnical Commission). It is a standard for compressing information of TV signals up to 5 Mbit / s, that is, about 1/100. In addition, the MPEG-1 standard sets the target quality to a medium quality that can be realized mainly at a transmission speed of about 1.5 Mbit / s, so that the MPEG standardized to meet the demand for higher image quality. -2 realizes TV broadcast quality at 2-15 Mbit / s for moving image signals. Furthermore, at present, the working group (ISO / IEC JTC1 / SC29 / WG11), which has been standardizing with MPEG-1 and MPEG-2, has achieved a compression ratio that exceeds MPEG-1 and MPEG-2, and further, on a per-object basis. MPEG-4, which enables encoding / decoding / operation and realizes new functions required in the multimedia era, has been standardized. In MPEG-4, it was originally aimed at standardizing a low bit rate encoding method, but now it has been extended to a more general purpose encoding including a high bit rate including interlaced images. Furthermore, at present, standardization activities of MPEG-4 AVC and ITU H.264 are advancing as a next-generation image encoding method with a higher compression rate jointly by ISO / IEC and ITU-T. As of August 2002, a so-called committee draft (CD) has been issued as a next-generation image coding system.

  In general, in encoding of moving images, the amount of information is compressed by reducing redundancy in the time direction and the spatial direction. Therefore, in inter-picture predictive coding for the purpose of reducing temporal redundancy, motion is detected and a predicted image is created in units of blocks with reference to the forward or backward picture, and the resulting predicted image and the encoded image are encoded. Encoding is performed on the difference value from the target picture. Here, a picture is a term representing a single screen, which means a frame in a progressive image and a frame or field in an interlaced image. Here, an interlaced image is an image in which one frame is composed of two fields having different times. In interlaced image encoding and decoding processing, one frame may be processed as a frame, processed as two fields, or processed as a frame structure or a field structure for each block in the frame. it can.

  A picture that is subjected to intra prediction encoding without using a reference picture is called an I picture. A picture for which inter-picture prediction coding is performed with reference to only one picture is called a P picture. In addition, a picture that can be subjected to inter-frame predictive coding with reference to two pictures at the same time is called a B picture. The B picture can refer to two pictures as an arbitrary combination of display times from the front or the rear. A reference picture (reference picture) can be specified for each block which is a basic unit of encoding and decoding. The reference picture described earlier in the encoded bitstream is designated as the first reference picture. The one described later is distinguished as the second reference picture. However, as a condition for encoding and decoding these pictures, the picture to be referenced needs to be already encoded and decoded.

  Motion compensation inter-picture prediction coding is used for coding a P picture or a B picture. The motion compensation inter-picture prediction encoding is an encoding method in which motion compensation is applied to inter-picture prediction encoding. Motion compensation is not simply predicting from the pixel value of the reference frame, but detecting the amount of motion of each part in the picture (hereinafter referred to as a motion vector) and performing prediction in consideration of the amount of motion. This improves the prediction accuracy and reduces the amount of data. For example, the amount of data is reduced by detecting the motion vector of the encoding target picture and encoding the prediction residual between the prediction value shifted by the motion vector and the encoding target picture. In the case of this method, since motion vector information is required at the time of decoding, the motion vector is also encoded and recorded or transmitted.

  The motion vector is detected in units of macroblocks. Specifically, the macroblock on the encoding target picture side is fixed, the macroblock on the reference picture side is moved within the search range, and is most similar to the reference block. The motion vector is detected by finding the position of the reference block.

  FIG. 1 is an explanatory diagram showing an example of the data structure of a bitstream. As shown in FIG. 1, the bit stream has the following hierarchical structure. The bit stream (Stream) is composed of a plurality of group of pictures. By using the group of pictures as the basic unit of encoding processing, editing of moving images and random access are possible. The group of pictures is composed of a plurality of pictures, and each picture includes an I picture, a P picture, or a B picture. Each picture is further composed of a plurality of slices. A slice is a band-like area in each picture, and is composed of a plurality of macroblocks. Streams, GOPs, pictures, and slices are further composed of a synchronization signal (sync) indicating the division of each unit and a header that is data common to the unit.

  In addition, when the stream is transmitted not by a continuous bit stream but by a packet or the like which is a unit of chopped data, the header part and the data part other than the header may be separated and transmitted separately. In that case, the header part and the data part do not become one bit stream as shown in FIG. However, in the case of a packet, even if the transmission order of the header part and the data part is not continuous, only the header part corresponding to the corresponding data part is transmitted in another packet, and it becomes one bit stream. Even if not, the concept is the same as that of the encoded bit stream described in FIG.

  In general, human visual characteristics are sensitive to low frequency components in an image, and high frequency components are said to be less sensitive than low frequency components. Furthermore, since the energy of the low frequency component of the image signal is larger than the energy of the high frequency component, the image encoding is performed in the order of the low frequency component to the high frequency component. As a result, the number of bits required for encoding the high frequency component is larger than the number of bits required for encoding the low frequency component.

  From the above viewpoint, in the conventional encoding method, in the quantization of the transform coefficient for each frequency obtained by orthogonal transform, the quantization step corresponding to the high frequency component is made rougher than the low frequency component. As a result, a method has been conventionally employed in which the compression rate is greatly improved while the subjective image quality is hardly deteriorated.

  The degree of coarsening of the high frequency component quantization step relative to the low frequency component depends on the image signal. Therefore, a method is adopted in which the size of each frequency component quantization step is changed according to the image. Has been. A quantization matrix is used to derive the quantization step for each frequency component. An example of the quantization matrix is shown in FIG. In the example of the quantization matrix in FIG. 2, the upper left corresponds to the DC component, the right corresponds to the horizontal high frequency component, and the lower portion corresponds to the vertical high frequency component. Moreover, it shows that a quantization step becomes coarse, so that a numerical value is large. The quantization matrix can usually be changed for each picture. A value indicating the magnitude of the quantization step of each frequency component is encoded with a fixed length. Note that each component of the quantization matrix and the value of the quantization step are generally in a proportional relationship, but it is not always necessary to stick to this, and it is sufficient that the correspondence between the two is clearly defined.

  However, since the value of each frequency component of the quantization matrix is concentrated in a certain range, there is a problem that the encoding efficiency is not good if the fixed length encoding is simply performed.

  An object of the present invention is to provide an image encoding method, an image decoding method, an image encoding device, an image decoding device, and a program that improve the encoding efficiency of a quantization matrix.

  In order to achieve the above object, the image encoding method of the present invention encodes an image by orthogonal transform and quantization in units of blocks, and performs quantization to derive a quantization step corresponding to each frequency of the orthogonal transform coefficient. An image encoding method for encoding a matrix, wherein a difference value between a frequency component and a predetermined value corresponding to the frequency component is obtained for each frequency component of the quantization matrix, and the difference value is encoded into a variable length code. In other words, the variable length code has a code length that is shorter or the same as the difference value is smaller. Here, the predetermined value may be, for example, the value of the frequency component having the highest appearance frequency, a constant indicating the average value of the frequency components, or a value determined in advance for each frequency component.

  According to this configuration, since the difference value between the frequency component and the predetermined value is taken, the difference value is smaller than the frequency component, and the difference value is encoded instead of encoding the frequency component itself. It is possible to shorten the code length of the long code and improve the coding efficiency.

  Here, the predetermined value may be a value of a frequency component corresponding to the difference value obtained immediately before.

  According to this configuration, since the frequency component and the immediately preceding frequency component are usually correlated, each difference value can be made smaller, and the encoding efficiency can be further improved.

  Here, the difference value may be obtained in order from a low frequency to a high frequency of frequency components included in the quantization matrix.

  According to this configuration, since frequency components are encoded in order from low frequency to high frequency, there is a high probability that the frequency component and the immediately preceding frequency component have the same value, and each difference value is more reliably reduced. Encoding efficiency can be further improved.

  Here, the difference value may be obtained as a remainder obtained by dividing by 2 to the power of k (k is a constant).

  According to this configuration, for example, when the difference value is expressed as a remainder divided by 2 to the 8th power (k = 8), the difference value can be reduced to a value that can be substantially expressed by 8 bits. Further, the number of bits of the variable length code can be reduced.

  Here, the image encoding method further determines whether or not a continuous difference value 0 exists at the end of the quantization matrix, and determines that the continuous difference value 0 exists, and determines that the continuous difference value 0 exists. The value 0 may not be variable length encoded.

  Here, in the encoding method, an end code may be added at the end of the variable length encoding to indicate that the subsequent difference values are 0 that are not encoded.

  According to this configuration, when a plurality of frequency components having the same value are continuous at the end of the quantization matrix, a plurality of variable length codes having the same bit string are not generated, but a plurality of frequency components are generated. Since only one variable length code is generated, the encoding efficiency can be further improved. In this case, the efficiency of the decoding process of the image decoding apparatus can be improved by adding the end code.

  In order to achieve the above object, the image decoding method of the present invention decodes an encoded image by inverse quantization and inverse orthogonal transform in units of blocks, and decodes an encoded quantization matrix. An image decoding method, wherein a variable-length-encoded quantization matrix is variable-length decoded into a difference value for each frequency component, and the difference value and a predetermined value corresponding to the frequency component are added, The frequency component of the quantization matrix is obtained, and the variable length code has a code length that is shorter or the same as the difference value is smaller.

  Here, the predetermined value may be a value of a frequency component obtained by the previous addition.

  Here, the order of the addition may be from the low frequency to the high frequency of the frequency component.

  Here, the frequency component may be obtained as a remainder divided by 2 to the power of k (k is a constant).

  According to this configuration, it is possible to easily decode a variable-length code string whose encoding efficiency has been improved by the above encoding method.

  Here, in the image decoding method, when the end code is detected from the variable-length-coded quantization matrix, the value of each subsequent frequency component is the same as the frequency component immediately before the end code. A value may be output.

  According to this configuration, when a plurality of frequency components having the same value are continuous at the end of the quantization matrix, the plurality of frequency components can be decoded from one variable length code. Further, it can be easily determined by the end code that a plurality of frequency components having the same value are continuous at the end of the quantization matrix.

  Also, the image encoding device, the image decoding device, and the program of the present invention have the same configuration and effect as described above.

  Hereinafter, Embodiment 1 of the present invention will be described with reference to FIGS.

  FIG. 3 is a block diagram showing the configuration of the image coding apparatus according to the first embodiment.

  The moving image encoding apparatus 1 is an apparatus that outputs an image encoded signal Str obtained by compressing and encoding an input image signal Vin into a bit stream such as a variable length code, and includes a motion detection unit ME and a motion compensation unit MC. A subtracting unit Sub, an orthogonal transform unit T, a quantization unit Q, an inverse quantization unit IQ, an inverse orthogonal transform unit IT, an addition unit Add, a picture memory PicMem, a switch SW, and a variable length coding unit VLC.

  The image signal Vin is input to the subtraction unit Sub and the motion detection unit ME. The subtraction unit Sub calculates a difference value between the input image signal Vin and the predicted image and outputs the difference value to the orthogonal transform unit T. The orthogonal transform unit T converts the difference value into a frequency coefficient and outputs it to the quantization unit Q.

  The quantization unit Q quantizes the block-unit frequency coefficient input to the subtraction unit Sub in a quantization step derived with reference to the quantization matrix WM input from the outside, and the quantized value Qcoef to the variable-length encoding unit. Output to.

  The inverse quantization unit IQ inversely quantizes the quantized value Qcoef in a quantization step derived with reference to the quantization matrix WM, restores it to a frequency coefficient, and outputs it to the inverse orthogonal transform unit IT.

  The inverse orthogonal transform unit IT performs inverse frequency conversion from the frequency coefficient to the pixel difference value, and outputs the result to the addition unit Add. The addition unit Add adds the pixel difference value and the predicted image output from the motion compensation unit MC to obtain a decoded image. The switch SW is turned ON when an instruction to store the decoded image is given, and the decoded image is stored in the picture memory PicMem.

  On the other hand, the motion detection unit ME to which the image signal Vin is input in units of macroblocks uses the decoded image stored in the picture memory PicMem as a search target and detects the position by detecting the image region closest to the input image signal. Is determined. Motion vector detection is performed in units of blocks obtained by further dividing a macroblock. At this time, since a plurality of pictures can be used as reference pictures, an identification number (reference index Index) for designating a picture to be referenced is required for each block. The reference picture can be specified by taking correspondence with the picture number of each picture in the picture memory PicMem by the reference index Index.

  In the motion compensation unit MC, using the motion vector detected by the above process and the reference index Index, an image region optimal for the predicted image is extracted from the decoded image stored in the picture memory PicMem.

  The variable length coding unit VLC performs variable length coding on the quantization matrix WM, the quantized value Qcoef, the reference index Index, and the motion vector MV to obtain a coded stream Str. Therefore, the variable length coding unit VLC has a first coding unit and a second coding unit inside. A first encoding unit (hereinafter referred to as a WM encoding unit) performs variable length encoding on the quantization matrix WM. The second encoding unit performs variable-length encoding on the quantization value Qcoef, the reference index Index, the motion vector MV, and the like other than the quantization matrix WM. The WM encoding unit obtains a difference value between the frequency component and a predetermined value corresponding to the frequency component for each frequency component of the quantization matrix, and encodes the difference value into a variable length code. This variable length code has a shorter code length as the difference value is smaller except for a part.

  FIG. 4 is a block diagram illustrating a configuration of the WM encoding unit. As shown in the figure, the WM encoding unit includes a subtracter 41, a W buffer 42, an offset determination unit 43, an adder 44, a block buffer 45, a number determination unit 46, an end code holding unit 47, a switch 48, and a variable length encoding. A portion 49 is provided.

  The subtractor 41 obtains a difference value between the frequency component and a predetermined value corresponding to the frequency component for each frequency component of the quantization matrix. The predetermined value may be, for example, the value of the frequency component having the highest appearance frequency, a constant indicating the average value of the frequency components, or a value predetermined for each frequency component. In the present embodiment, the predetermined value is the value of the frequency component corresponding to the difference value obtained immediately before. In this case, the subtractor 41 obtains a difference value between the current frequency component W [i] input from the outside and the previous frequency component W [i−1] held in the W buffer 42. Since the frequency component W [i] and the immediately preceding frequency component W [i-1] are normally correlated, each difference value can be set to a smaller value. W [i] represents the i-th frequency component in the coding order.

  5A to 5D are diagrams illustrating an example of the order of the frequency components of the quantization matrix input to the subtractor 41. FIG. This order is the scan order of the quantization matrix input to the subtracter 41. In the orthogonal transform of image coding, two types of 4 × 4 pixel units and 8 × 8 pixel units are most often used. 5A and 5B show an example of a 4 × 4 pixel unit, and FIGS. 5C and 5D show an example of an 8 × 8 pixel unit. In this embodiment, each component of the quantization matrix is individually encoded in any of FIGS. 8A to 8C, so that the components are encoded in the order of low frequency components to high frequency components as shown in FIGS. 5A and 5C. However, there is no difference in encoding efficiency even if encoding is simply performed in the horizontal order as shown in FIGS. 5B and 5B.

  FIG. 6A shows a part of the frequency component of the quantization matrix input to the subtractor 41. W [0], W [1], W [2], W [3],... In FIG.

  The W buffer 42 is a buffer that holds the immediately preceding frequency component W [i-1]. Before the start of encoding, the W buffer 42 holds a typical value that can be taken most as a DC component as an initial value W [−1]. In the present embodiment, the initial value W [−1] is 8. Note that the W buffer 42 may hold a predetermined value K corresponding to the frequency component in advance instead of holding the immediately preceding frequency component W [i−1].

  The offset determining unit 43 determines an offset value for converting the difference (W [i] −W [i−1]) from the subtractor 41 into a remainder obtained by dividing the difference by 2 to the power of k (here, k = 8). To do. That is, a value of −256, 0, or 256 is output so that the result of adding the offset to (W [i] −W [i−1]) becomes a value from −128 to 127.

  The adder 44 calculates the difference value D [i] by adding the difference value (W [i] −W [i−1]) from the subtractor 41 and the offset value from the offset determination unit 43. Note that D [i] represents the i-th difference value in the encoding order.

  FIG. 6B shows an example of the difference value D [i] calculated by the adder 44. Difference values D [0], D [1], D [2], D [3]... In FIG. 6 are the frequency components W [0], W [1], W [2], W in FIG. [3] is an example corresponding to.

  The block buffer 45 is a buffer that holds the difference value D [i] input from the adder 44 for one block (one quantization matrix).

  The number determination unit 46 counts the number of difference values 0 (W [i] −W [i−1] = 0) continuously present at the end of the scan order (the end of the block), and holds it in the block buffer 45. 48 is controlled to output D [0] to D [M]. Here, D [M] is a difference value immediately before the first difference value 0 of the continuously existing difference value 0.

  The end code holding unit 47 holds an end code indicating the end of the variable length code of the quantization matrix. The end code may be a value that cannot be taken by the difference value D [i].

  The switch 48 selects the output of the block buffer 45 so as to output D [0] to D [M] held in the block buffer 45, and then outputs the end code holding unit 47 so as to output one end code. Select. That is, the output of the switch 48 is one end code from the difference value D [0] corresponding to the head of the quantization matrix to the difference value D [M] immediately before the difference value 0 that is continuously present last. is there. When the number of difference values 0 (W [i] −W [i−1] = 0) continuously present at the end of the scan order (the end of the block) counted by the number determination unit 46 is 0. In this case, the switch 48 does not output an end code.

  The variable length coding unit 49 performs variable length coding on the difference value input from the block buffer 45 via the switch 48 and the end code so that the smaller the difference value, the shorter or the same code length. To do.

  FIG. 6C shows a specific example of a code encoded by the variable length encoding unit 49 using the code of FIG. 8C. In the figure, W [0] to W [4] and D [0] to D [4] correspond to FIGS. 6A and 6B. The code by the variable length encoding unit 49 is, for example, a code length of 1 bit when the difference value is 0, and a code length of 3 bits when the difference value is 1. Thus, the variable length code by the variable length encoding unit 49 has a shorter code length as the difference value is smaller. The frequency components W [30] to W [63] all have the same value 64. In this case, the difference values D [30] to D [63] are continuously 0. In this case, the difference values D [30] to D [63] are not encoded. Instead of the code “1” for 0 of the difference value D [30], the end code is variable-length encoded.

  FIG. 7 is a flowchart showing detailed encoding processing of the quantization matrix in the WM encoding unit. In the figure, Num is the number of frequency components (16, 64, etc.) in the quantization matrix, and i and j are variables for counting from 0 to (Num-1), respectively.

  The WM encoding unit first holds the initial value W [−1] (for example, 8) in the W buffer 42 (S71), and performs a difference for each frequency component of the quantization matrix in the process shown in loop 1 (S72 to S77). A value is obtained and stored in the block buffer 45, and then variable length coding of the difference value is performed by the processing shown in loop 2 (S78 to S81).

  In the loop 1, the subtractor 41 obtains a difference value D between the current frequency component W [i] input from the outside and the previous frequency component W [i-1] held in the W buffer 42 (S73). ). The offset determination unit 43 determines an offset value for converting the difference D from the subtractor 41 into a remainder obtained by dividing 2 by the k power (here, k = 8) (S74). For example, when W [i] can take a value of −256 to +254, the offset determining unit 43 sets the offset value to −256 when D is 128 or more, and the offset value is 256 and D when D is less than −128. Otherwise, the offset value is determined to be 0. Thereby, D after adding the offset value (that is, D [i]) becomes a remainder including not only a positive value but also a negative value within a range of −128 to +127. As a result, the value that can be taken by the difference value D [i] is reduced, so that the variable length code of D [i] when encoded using the code of FIG.

  The adder 44 calculates the difference value D [i] by adding the difference value D (= W [i] −W [i−1]) from the subtractor 41 and the offset value from the offset determination unit 43. (S75-76).

  In the loop 2, the number determination unit 46 determines the number of difference values 0 that are continuously present at the end of the D [i] column (S79), and the difference value D [M] immediately before the consecutive difference value 0 is determined. The switch 48 is connected to the output of the block buffer 45 and then connected to the output of the end code holding unit 47.

  The variable length encoding unit 49 encodes the difference value D [i] input from the block buffer 45 via the switch 48 and the end code input from the end code holding unit 47 from the switch 48 (S82). The end code may be any value that cannot be taken by D [i]. Here, -W [M], that is, W [M] = 0. Since the frequency component of the quantization matrix is a positive number, the fact that W [M] is 0 can be identified as an end code.

  8A to 8C are explanatory diagrams illustrating an example of variable length coding in the variable length coding unit 49. Both use Exponential Golom code.

FIG. 8A shows a first example of a variable length code. In the same figure, the relationship between a code word (code) and the difference value (value) before encoding is shown. This example can be used when the value that the difference value can take is a positive number. The larger the difference value, the lower the occurrence frequency and the longer the code length. Therefore, a short code word is assigned to a small difference value. The conversion between the code word and the value of each component can be easily derived by the arithmetic expression shown in FIG. 8A. For example, when the difference value is 7, it is expressed as 7 = 4 + 3 = (square of 2 + binary number “11”). The code word code corresponding to this is a bit string in which “000 (three 0s)”, “1”, and “11” are arranged. N in FIG. 8A is the nearest power of 2 that does not exceed value. X ₀ X ₁ X ₂ ... X _N-1 is a bit string indicating a value obtained by subtracting the nearest power of 2 from value.

  FIG. 8B shows a second example of the variable length code. In the figure, the frequency column W [i] (that is, the predetermined value K + difference value) of the quantization matrix is written in the value column. This is suitable when the value of the component W [i] of the quantization matrix is most likely to take the predetermined value K. Since the occurrence frequency of the predetermined value K is high, the code length is the shortest when the component value is K, and the code length is long when the component value is away from K. The predetermined value K may be W [i-1], for example, but may be a predetermined constant.

  FIG. 8C shows a third example of variable length coding. In the same figure, the relationship between a code word (code) and the difference value (value) before encoding is shown. This example can be used when the possible value of the difference value is not only a positive number but also a negative number. When the components of the quantization matrix are encoded in the order shown in FIGS. 5A to 5D, the values of the adjacent frequency components W [i−1] have a large correlation. Therefore, when encoding in any order of FIGS. 5A to 5D, the difference value is concentrated in the vicinity of 0 by encoding the difference value with the value W [i−1] of the component encoded immediately before. To do. As a result, as shown in FIG. 8C, the number of bits of the variable-length code can be further reduced by using a codeword having a short code length and a code code having a long code length for encoding values near 0.

  9A to 9C are diagrams illustrating a data structure of a quantization matrix and a quantization matrix encoded in a stream. In the figure, Header corresponds to the stream, GOP, or picture header of FIG. FIG. 9A is a diagram illustrating an arrangement of frequency components of the quantization matrix. Wi, j represents a component of a quantization matrix of i rows and j columns. 9B and 9C show how the encoded data of each component of the quantization matrix is arranged in the header portion. A bit stream obtained by encoding a quantization matrix is WeightingMatrix. FIG. 9B shows a stream encoded in the order of FIG. 5C, and FIG. 9C shows a stream encoded in the order of FIG. 5D. Note that “Wi, j” in the streams of FIGS. 9B and 9C means an encoded variable-length code corresponding to Wi, j.

  As described above, according to the WM encoding unit in the present embodiment, since the difference value D [i] between the frequency component W [i] and the predetermined value K is taken, the difference value D [i] is the frequency component. A value smaller than W [i]. Since the W encoding unit encodes the difference value D [i] rather than encoding the frequency component W [i] itself, it is possible to shorten the code length of the variable length code and improve the encoding efficiency. it can.

  Further, by using the immediately preceding frequency component W [i-1] as the predetermined value K, each difference value D [i] can be made smaller, and the encoding efficiency can be further improved. This is because the frequency component W [i] and the immediately preceding frequency component W [i-1] are usually correlated.

  Further, since the frequency component W [i] is encoded in order from the low frequency to the high frequency, the frequency component W [i] and the immediately preceding frequency component W [i-1] have the same value. Therefore, each difference value D [i] can be more reliably reduced and the encoding efficiency can be further improved.

  Further, when the difference value D [i] is expressed as a remainder obtained by dividing by 2 to the 8th power (k = 8), the difference value can be reduced to a value that can be expressed substantially by 8 bits. The number of bits of the variable length code can be reduced.

  Also, when multiple frequency components with the same value are consecutive at the end of the quantization matrix, instead of generating multiple variable length codes with the same bit string, variable length codes for multiple frequency components Is not generated, the encoding efficiency can be further improved. In this case, the efficiency of the decoding process of the image decoding apparatus can be improved by adding the end code.

  FIG. 10 is a block diagram of the image decoding apparatus according to the first embodiment. In the figure, units that perform the same operations as those in the block diagram of the image coding apparatus in FIG.

  The variable length decoding unit VLD decodes the encoded stream Str, and outputs a quantization matrix WM, a quantization value Qcoef, a reference index Index, and a motion vector MV.

  Therefore, the variable length decoding unit VLD includes a first decoding unit and a second decoding unit inside. A first decoding unit (hereinafter referred to as WM decoding unit) performs variable length decoding on the encoded quantization matrix WM. The second decoding unit performs variable-length decoding on the encoded quantization value Qcoef, the reference index Index, the motion vector MV, and the like other than the encoded quantization matrix WM. The WM decoding unit variable-length-decodes the variable-length-encoded quantization matrix to a difference value for each frequency component, and adds the difference value and a predetermined value corresponding to the frequency component to obtain a quantization matrix Is obtained.

  The quantization matrix QM, the quantization value Qcoef, the reference index Index, and the motion vector MV are input to the picture memory PicMem, the motion compensation unit MC, and the inverse quantization unit IQ, and the decoding process is performed. This has already been described in the block diagram of the image decoding apparatus.

  FIG. 11 is a block diagram illustrating a configuration of the WM decoding unit. The WM decoding unit includes a variable length decoding unit 51, an adder 52, a remainder calculation unit 53, a switch 54, a W buffer 55, a switch 56, and a final coefficient determination unit 57.

  The variable length decoding unit 51 decodes the variable length code of the variable length encoded frequency component into a difference value D [i] for each frequency component.

  The adder 52 adds the difference value D [i] from the variable length decoding unit 51 and the value of the previous frequency component W [i−1] from the W buffer 55.

  The multiplier calculation unit 53 adds the 2k power to the addition result from the adder 52, and calculates the remainder of the 2k power to obtain the frequency component W [i]. Specifically, the multiplier calculation unit 53 calculates (D [i] + W [i-1] +256)% 256. Here,% indicates an operation for obtaining a remainder obtained by division (by 256).

  The switch 54 is turned on while the end code is not output from the remainder calculation unit 53, and is turned off when the end code is output.

  The W buffer 55 holds the frequency component W [i] input from the remainder calculation unit 53 via the switch 54, and outputs it to the adder 52 as the frequency component W [i-1] in the next cycle. The W buffer 55 holds W [i−1] (for example, 8) as an initial value.

  The switch 56 selects the frequency component W [i] input from the multiplier calculation unit 53 via the switch 54 while the end code is not output from the multiplier calculation unit 53, and the end code is output. After that, the W buffer 55 is selected.

  The final coefficient determination unit 57 determines whether an end code is output from the multiplier calculation unit 53 and controls the switch 54 and the switch 56 according to the determination result.

  As described above, W [i] corresponding to all frequency components can be decoded. Even if the end code is not detected, if the same number of difference values D [i] as all frequency components included in the block are decoded, the decoding of the quantization matrix is ended at that point.

  FIG. 12 is a flowchart showing the WM decoding process in the WM decoding unit. In the figure, Num is the number of frequency components in the quantization matrix (16, 64, etc.), i is a variable for counting from 0 to (Num-1), and j is counted from i to (Num-1). It is a variable to do. As shown in the figure, the WM decoding unit first holds the initial value W [i-1] in the W buffer 55 (S121), performs variable-length decoding and output in the loop 1 (S122 to 127), and the end code is After detection, continuous frequency coefficients having the same value are output in loop 2 (S128 to 130).

  In the loop 1, the variable length decoding unit 51 decodes the input variable length code into the difference value D [i] (S123). Further, the adder 52 adds the difference value from the variable length decoding unit 51 and the immediately preceding frequency component from the W buffer 55, and the remainder calculation unit 53 calculates the remainder W = (D [i] + W [i -1] +256)% 256 is calculated (S124). The final coefficient determination unit 57 determines whether or not the remainder W is an end code (here, whether or not it is positive) (S125). If it is determined that the code is not an end code, the remainder W is output as a frequency component W [i−1] from the multiplier calculation unit 53 via the switch 56 and is simultaneously held in the W buffer 55 via the switch 54. (S126). If it is determined that it is an end code, the process of loop 2 is executed.

  In the loop 2, when a plurality of frequency components having the same value are continuous at the end of the quantization matrix, the frequency component W [i-1] is continuously output as the frequency component W [i]. That is, W [i-1] held in the W buffer 55 is output via the switch 56 as each subsequent frequency component W [i] to W [Num-1] (S129).

  As described above, according to the WM decoding unit in the present embodiment, the variable length code differentially encoded by the WM encoding unit is correctly decoded. Furthermore, when a plurality of frequency components having the same value are continuous at the end of the quantization matrix, the WM decoding unit can decode the plurality of frequency components from one variable length code. At that time, it can be easily determined by the end code that a plurality of frequency components having the same value are continuous at the end of the quantization matrix.

  Furthermore, the program for realizing the image encoding method and the image decoding method shown in each of the above embodiments is recorded on a recording medium such as a flexible disk, so that the program shown in each of the above embodiments is shown. Processing can be easily performed in an independent computer system.

  FIGS. 13A to 13C are explanatory diagrams when the image coding method and the image decoding method according to each of the above embodiments are implemented by a computer system using a program recorded on a recording medium such as a flexible disk.

  FIG. 13B shows an appearance, a cross-sectional structure, and a flexible disk as viewed from the front of the flexible disk, and FIG. 13A shows an example of a physical format of the flexible disk that is a recording medium body. The flexible disk FD is built in the case F, and on the surface of the disk, a plurality of tracks Tr are formed concentrically from the outer periphery toward the inner periphery, and each track is divided into 16 sectors Se in the angular direction. ing. Therefore, in the flexible disk storing the program, the program is recorded in an area allocated on the flexible disk FD.

  FIG. 13C shows a configuration for recording and reproducing the program on the flexible disk FD. When the program for realizing the image encoding method and the image decoding method is recorded on the flexible disk FD, the program is written from the computer system Cs via the flexible disk drive. When the image encoding method and the image decoding method for realizing the image encoding method and the image decoding method by the program in the flexible disk are built in the computer system, the program is read from the flexible disk by the flexible disk drive. Transfer to computer system.

  In the above description, a flexible disk is used as the recording medium, but the same can be done using an optical disk. Further, the recording medium is not limited to this, and any recording medium such as an IC card or a ROM cassette capable of recording a program can be similarly implemented.

  Furthermore, application examples of the image coding method and the image decoding method shown in the above embodiment and a system using the same will be described.

  FIG. 14 is a block diagram showing an overall configuration of a content supply system ex100 that implements a content distribution service. The communication service providing area is divided into desired sizes, and base stations ex107 to ex110, which are fixed radio stations, are installed in each cell.

  The content supply system ex100 includes, for example, a computer ex111, a PDA (personal digital assistant) ex112, a camera ex113, a mobile phone ex114, a camera via the Internet ex101, the Internet service provider ex102, the telephone network ex104, and the base stations ex107 to ex110. Each device such as the attached mobile phone ex115 is connected.

  However, the content supply system ex100 is not limited to the combination as shown in FIG. 14, and any of the combinations may be connected. Further, each device may be directly connected to the telephone network ex104 without going through the base stations ex107 to ex110 which are fixed wireless stations.

  The camera ex113 is a device capable of shooting a moving image such as a digital video camera. The mobile phone is a PDC (Personal Digital Communications) system, a CDMA (Code Division Multiple Access) system, a W-CDMA (Wideband-Code Division Multiple Access) system, or a GSM (Global System for Mobile Communications) system mobile phone, Alternatively, PHS (Personal Handyphone System) or the like may be used.

  Further, the streaming server ex103 is connected from the camera ex113 through the base station ex109 and the telephone network ex104, and live distribution based on the encoded data transmitted by the user using the camera ex113 can be performed. The encoded processing of the captured data may be performed by the camera ex113 or may be performed by a server or the like that performs data transmission processing. The moving image data shot by the camera 116 may be transmitted to the streaming server ex103 via the computer ex111. The camera ex116 is a device that can shoot still images and moving images, such as a digital camera. In this case, the encoding of the moving image data may be performed by the camera ex116 or the computer ex111. The encoding process is performed in the LSI ex117 included in the computer ex111 and the camera ex116. Note that image encoding / decoding software may be incorporated in some storage medium (CD-ROM, flexible disk, hard disk, etc.) that is a recording medium readable by the computer ex111 or the like. Furthermore, you may transmit moving image data with the mobile telephone ex115 with a camera. The moving image data at this time is data encoded by the LSI included in the mobile phone ex115.

  In this content supply system ex100, the content (for example, video shot of music live) captured by the user with the camera ex113, camera ex116, etc. is encoded and transmitted to the streaming server ex103 as in the above embodiment. On the other hand, the streaming server ex103 distributes the content data to the requested client. Examples of the client include a computer ex111, a PDA ex112, a camera ex113, a mobile phone ex114, and the like that can decode the encoded data. In this way, the content supply system ex100 can receive and reproduce the encoded data at the client, and also realize personal broadcasting by receiving, decoding, and reproducing in real time at the client. It is a system that becomes possible.

  The image encoding device or the image decoding device described in the above embodiments may be used for encoding and decoding of each device constituting this system.

  A mobile phone will be described as an example.

  FIG. 15 is a diagram showing a mobile phone ex115 using the image coding method and the image decoding method described in the above embodiment. The cellular phone ex115 includes an antenna ex201 for transmitting and receiving radio waves to and from the base station ex110, a camera such as a CCD camera, a camera unit ex203 capable of taking a still image, a video shot by the camera unit ex203, and an antenna ex201. A display unit ex202 such as a liquid crystal display for displaying data obtained by decoding received video and the like, a main body unit composed of a group of operation keys ex204, an audio output unit ex208 such as a speaker for audio output, and audio input To store encoded data or decoded data such as a voice input unit ex205 such as a microphone, captured video or still image data, received mail data, video data or still image data, etc. Recording medium ex207, and slot portion ex206 for enabling the recording medium ex207 to be attached to the mobile phone ex115. Have. The recording medium ex207 stores a flash memory element which is a kind of EEPROM (Electrically Erasable and Programmable Read Only Memory) which is a nonvolatile memory that can be electrically rewritten and erased in a plastic case such as an SD card.

  Further, the cellular phone ex115 will be described with reference to FIG. The mobile phone ex115 controls the power supply circuit ex310, the operation input control unit ex304, and the image encoding for the main control unit ex311 configured to control the respective units of the main body unit including the display unit ex202 and the operation key ex204. Unit ex312, camera interface unit ex303, LCD (Liquid Crystal Display) control unit ex302, image decoding unit ex309, demultiplexing unit ex308, recording / reproducing unit ex307, modulation / demodulation circuit unit ex306, and audio processing unit ex305 via a synchronization bus ex313 Are connected to each other.

  When the end call and power key are turned on by a user operation, the power supply circuit ex310 activates the camera-equipped digital mobile phone ex115 by supplying power from the battery pack to each unit. .

  The mobile phone ex115 converts the voice signal collected by the voice input unit ex205 in the voice call mode into digital voice data by the voice processing unit ex305 based on the control of the main control unit ex311 including a CPU, a ROM, a RAM, and the like. The modulation / demodulation circuit unit ex306 performs spread spectrum processing, the transmission / reception circuit unit ex301 performs digital analog conversion processing and frequency conversion processing, and then transmits the result via the antenna ex201. In addition, the cellular phone ex115 amplifies the received signal received by the antenna ex201 in the voice call mode, performs frequency conversion processing and analog-digital conversion processing, performs spectrum despreading processing by the modulation / demodulation circuit unit ex306, and analog audio by the voice processing unit ex305. After being converted to a signal, this is output via the audio output unit ex208.

  Further, when an e-mail is transmitted in the data communication mode, text data of the e-mail input by operating the operation key ex204 on the main body is sent to the main control unit ex311 via the operation input control unit ex304. The main control unit ex311 performs spread spectrum processing on the text data in the modulation / demodulation circuit unit ex306, performs digital analog conversion processing and frequency conversion processing in the transmission / reception circuit unit ex301, and then transmits the text data to the base station ex110 via the antenna ex201.

  When transmitting image data in the data communication mode, the image data captured by the camera unit ex203 is supplied to the image encoding unit ex312 via the camera interface unit ex303. When image data is not transmitted, the image data captured by the camera unit ex203 can be directly displayed on the display unit ex202 via the camera interface unit ex303 and the LCD control unit ex302.

  The image encoding unit ex312 has a configuration including the image encoding device described in the present invention, and an encoding method using the image data supplied from the camera unit ex203 in the image encoding device described in the above embodiment. The encoded image data is converted into encoded image data by compression encoding, and sent to the demultiplexing unit ex308. At the same time, the cellular phone ex115 sends the sound collected by the audio input unit ex205 during imaging by the camera unit ex203 to the demultiplexing unit ex308 as digital audio data via the audio processing unit ex305.

  The demultiplexing unit ex308 multiplexes the encoded image data supplied from the image encoding unit ex312 and the audio data supplied from the audio processing unit ex305 by a predetermined method, and the multiplexed data obtained as a result is a modulation / demodulation circuit unit A spectrum spread process is performed at ex306, a digital-analog conversion process and a frequency conversion process are performed at the transmission / reception circuit unit ex301, and then transmitted through the antenna ex201.

  When receiving data of a moving image file linked to a home page or the like in the data communication mode, the received signal received from the base station ex110 via the antenna ex201 is subjected to spectrum despreading processing by the modulation / demodulation circuit unit ex306, and the resulting multiplexing is obtained. Is sent to the demultiplexing unit ex308.

  In addition, in order to decode the multiplexed data received via the antenna ex201, the demultiplexing unit ex308 separates the multiplexed data, thereby encoding the encoded bit stream of the image data and the encoded bit stream of the audio data. The encoded image data is supplied to the image decoding unit ex309 via the synchronization bus ex313, and the audio data is supplied to the audio processing unit ex305.

  Next, the image decoding unit ex309 is configured to include the image decoding device described in the present invention, and a decoding method corresponding to the encoding method described in the above embodiment for an encoded bit stream of image data. To generate playback moving image data, which is supplied to the display unit ex202 via the LCD control unit ex302, thereby displaying, for example, moving image data included in the moving image file linked to the homepage . At the same time, the audio processing unit ex305 converts the audio data into an analog audio signal, and then supplies the analog audio signal to the audio output unit ex208. Thus, for example, the audio data included in the moving image file linked to the home page is reproduced. The

  It should be noted that the present invention is not limited to the above-described system, and recently, digital broadcasting using satellites and terrestrial waves has become a hot topic. As shown in FIG. Any of the decoding devices can be incorporated. Specifically, in the broadcasting station ex409, the encoded bit stream of the video information is transmitted to the communication or broadcasting satellite ex410 via radio waves. Receiving this, the broadcasting satellite ex410 transmits a radio wave for broadcasting, and receives the radio wave with a home antenna ex406 having a satellite broadcasting receiving facility, such as a television (receiver) ex401 or a set top box (STB) ex407. The device decodes the encoded bit stream and reproduces it. In addition, the image decoding apparatus described in the above embodiment can also be implemented in a playback apparatus ex403 that reads and decodes an encoded bitstream recorded on a storage medium ex402 such as a CD or DVD as a recording medium. is there. In this case, the reproduced video signal is displayed on the monitor ex404. Further, a configuration in which an image decoding device is mounted in a set-top box ex407 connected to a cable ex405 for cable television or an antenna ex406 for satellite / terrestrial broadcasting, and this is reproduced on the monitor ex408 of the television is also conceivable. At this time, the image decoding apparatus may be incorporated in the television instead of the set top box. It is also possible to receive a signal from the satellite ex410 or the base station ex107 by the car ex412 having the antenna ex411 and reproduce the moving image on a display device such as the car navigation ex413 that the car ex412 has.

Further, the image signal can be encoded by the image encoding device shown in the above embodiment and recorded on a recording medium. Specific examples include a recorder ex420 such as a DVD recorder that records image signals on a DVD disk ex421 and a disk recorder that records on a hard disk. Further, it can be recorded on the SD card ex422. If the recorder ex420 includes the image decoding device described in the above embodiment, the image signal recorded on the DVD disc ex421 or the SD card ex422 can be reproduced and displayed on the monitor ex408.
For example, the configuration of the car navigation ex413 may be a configuration excluding the camera unit ex203, the camera interface unit ex303, and the image encoding unit ex312 in the configuration illustrated in FIG. 16, and the same applies to the computer ex111 and the television (receiver). ) Ex401 can also be considered.

  In addition to the transmission / reception type terminal having both the encoder and the decoder, the terminal such as the mobile phone ex114 has three mounting formats: a transmitting terminal having only an encoder and a receiving terminal having only a decoder. Can be considered.

  Each of the functional blocks in the block diagrams shown in FIGS. 3, 4, 10, and 11 is typically realized as an LSI that is an integrated circuit device. This LSI may be made into one chip or a plurality of chips. (For example, the functional blocks other than the memory may be integrated into one chip.) Although the LSI is used here, it may be referred to as an IC, a system LSI, a super LSI, or an ultra LSI depending on the degree of integration.

  The method of circuit integration is not limited to LSI, and may be realized by a dedicated circuit or a general-purpose processor. An FPGA (Field Programmable Gate Array) that can be programmed after manufacturing the LSI or a reconfigurable processor that can reconfigure the connection and setting of circuit cells inside the LSI may be used.

  Further, if integrated circuit technology comes out to replace LSI's as a result of the advancement of semiconductor technology or a derivative other technology, it is naturally also possible to carry out function block integration using this technology. Biotechnology can be applied.

  Further, among the functional blocks, only the unit for storing data may be configured separately as in the recording medium 115 of the present embodiment, without being integrated into one chip.

  In the functional blocks of the block diagrams shown in FIGS. 3, 4, 10, and 11 and the flowcharts shown in FIGS. 7 and 12, the central part is also realized by a processor and a program.

  As described above, the image encoding method and the image decoding method described in the above embodiment can be used in any of the above-described devices / systems, and by doing so, the effects described in the above embodiment can be obtained. Obtainable.

  Hereinafter, Embodiment 2 of the present invention will be described.

  The present invention enables quantization of a matrix in a moving picture coding system using intra prediction, and enables expression of a quantized matrix. In particular, the present invention relates to generation of cyclic difference values suitable for signed exponential Golomb codes.

  Conventionally, quantization matrices have been used in moving picture coding, and it is also known that Exp-Golomb codes can be applied. Conventionally, in moving picture compression processing, a moving picture code is represented by a syntax structure having a hierarchical structure with corresponding semantics, and uses lossy coding and lossless coding, or variable length coding. Furthermore, transform domain quantization is also known and has been utilized.

  Here, a method and program for expressing, executing, and enabling moving image compression using a quantization weighting matrix will be described. The features of the present invention are as follows.

  1. It is possible to transmit and represent a quantization matrix in a moving picture coding system using intra-picture coding.

  2. Define and execute a quantization weighting matrix based on integers and realized by operations that do not require division.

  3. The quantization matrix is subjected to cyclic differential code and exponential Golomb coding that can increase the coding efficiency.

  4. A redundant weight value is not transmitted as communication transmission data, but is expressed as a quantization matrix that can be derived on the decoder side.

0. Difference encoding of quantization matrix In order to encode each component of the quantization matrix, the matrix component is first mapped to a one-dimensional array. This mapping uses a zigzag pattern described in MPEG-4 Part 10 (MPEG-2 and MPEG-4 Part 2 have similar definitions). Examples of zigzag patterns are shown in FIGS. The one-dimensional array obtained by the transformation includes 64 elements for an 8 × 8 quantization matrix and 16 elements for a 4 × 4 quantization matrix. According to the definition of the above standard, the included element is a positive integer between 1 and 255.

  In conventional differential encoding, a differential code is obtained by subtracting the value of the element immediately preceding each array element from each one-dimensional array element. However, 8 is subtracted from the head array element (because the value of the head element is a DC weight and usually takes a value of 8). Subsequently, each differential code is further encoded using an Exp-Golomb code as defined in MPEG-4 Part 10. (lSO / IEC JTC1 / SC29WG11, 14496-10).

  The conventional differential encoding method of the quantization matrix has the following problems.

  1. The difference between each value in the quantization matrix and the value immediately before the value is a value in the range of -254 to +254. When the range of values that can be taken by the difference value increases, the efficiency of variable length coding using Exp-Golomb becomes worse.

  2. When “truncating” is performed in order to reduce the number of matrix values to be encoded, a special value used to notify the end of a series of data needs to be an appropriate value. Otherwise, the “truncated” method is inappropriate.

  In the present invention, the above problem is solved by generating a cyclic difference value so that a range of values that can be taken is −128 to +128. This makes it possible to perform encoding more suitable for signed Exp-Golomb codes. The present invention can also solve the problem with the end of code sequence notification when using the “truncate” method.

  In the present invention, in order to improve the mechanism for notifying the end of consecutive codes while using the “truncated” method, the number of bits required for notifying the end is set to 9 bits or less. In order to realize this, a continuous code is suppressed using a plurality of signals. This is applicable to code arrays with 64 elements, but may also be applied to code arrays with 16 elements.

1. Inverse processing of cyclic differential encoding In this processing, D [i] is input and W [i] is output. Here, D [i] represents a differential code, and W [i] represents each value of the quantization matrix sampled by the zigzag scan.

  The following calculation is performed on the input W [i].

Define: M = 16, for 4x4 block; M = 64, for 8x8 block
Define M = 16, for 4x4 matrix or M = 64, for 8x8 matrix
W [0] = (D [0] + 8 + 256)% 256
k = 0;
for (i = 1; (i <M) && (W [i-1] + D [i] +256)% 256>0; i ++)
[
W [i] = (W [i-1] + D [i] +256)% 256
]
for (j = I; j <M; j ++) [
W [j] = W [i-1]
]

2. Differential code generation process In this process, W [i] is input and D [i] is output. Here, W [i] represents each value of the quantization matrix sampled by the zigzag scan, and D [i] represents a differential code obtained from W [i].

  The following calculation is performed on the input W [i].

Define M = 16, for 4x4 matrix or M = 64, for 8x8 matrix
D [0] = W [0] -8
k = M;
for (i = 1; i <M; i ++)
[
D [i] = W [i] -W [i-1]
if (D [i]> 128)
D [i] = D [i] -256
else if (D [i] <-128)
D [i] = D [i] +256
else if (D [i] == 0) && (k == M)
k = i
else if (D [i]! = 0)
k = M
]
// This is intended to insert an end signal D [k] that satisfies W [k] = 0
if (k! = M) [
if (W [k-1]> 128)
D [k] = 256-W [k-1]
else
D [k] = -W [k-1]
]
// This is for optimization purposes
if (Length_of-ExpGlomb_code (D [k])> Mki) [
for (i = k; i <M; i ++)
D [k] = 0
]

  Through the above calculation, a difference value D [i] having an absolute value of 128 or less is generated including a value indicating the end of the difference code string. Therefore, the differential code obtained by the above calculation is more suitable for signed Exp-Golomb coding.

3. Mechanism for notifying the end of the differential code sequence Instead of using signal values such as W and D to indicate the end of the differential code sequence, the present invention uses two or more codes, Let the encoder choose from the inside.

  In order to notify the end of the code string, the encoder can make the following settings. Here, the decoder needs to be able to recognize the settings made by the encoder.

  Assume that the end of the code string to be encoded is at k (that is, W [k] = W [k−1]) (W [i] takes a value from 0 to M. M Is 16 for 4 × 4 transform blocks and 64 for 8 × 8 transform blocks).

  − Set the value of W [k] to 0. This corresponds to the difference value D [k]. The absolute value of D [k] is 128 or less.

  − Set the value of D [k] to a value greater than 255 or less than -255. As a result, a signed Exp-Golomb code 0000000001xxxxxxxxx in which the first 9 bits are 0 is generated. The encoder need only output the first 10 bits (here, it is also possible to output only 9 bits without outputting “1”). After the decoding process, the decoder obtains a value exceeding 255 as a signed Golomb code. If the decoder is capable of handling an illegal Golomb code, it is only necessary to output 9 consecutive 0s.

  Corresponding to this, the decoding process is performed in the following procedure. (In the pseudo code given below, instead of checking whether D [k] takes a value of +255 to −255, it simply checks whether D [k] is a value within an appropriate range. If D [k] does not take a value within the appropriate range, it is determined that D [k] is the end of a continuous code.In each case, a check is necessary, so two conditional operations are omitted. can do.)

Define: M = 16, for 4x4 block; M = 64, for 8x8 block
Define M = 16, for 4x4 matrix or M = 64, for 8x8 matrix
D [0] = (W [0] + 8 + 256)% 256
k = 0;
for (i = 1; (i <M) && (Golombcode Legal) && (-129 <D [i] <129) && (W [i-1] +256)% 256>0; i ++)
W [i] = (W [i-1] + D [i] +256)% 256
[
for (j = i; j <M; j ++) [
W [j] = W [i-1]
]

4. Summary In order to improve coding efficiency, a fixed-length code and an Exp-Golomb code are used in combination as necessary.

− Generation of cyclic difference − 2-bit (or 4-bit) LSB with fixed-length code
− 6-bit (or 4-bit) MSB using Exp-Golomb code
By using these, the total number of bits can be further reduced.

  The above "1. Inverse processing of cyclic differential encoding" can also be expressed as a quantization weighting matrix derivation process as follows.

  In this process, the value of the continuous quantization weighting matrix subjected to Exp-Golomb encoding is input, and the value of the quantization weighting matrix expressed as a two-dimensional array is output.

  -Invoking the decoding process to decode the values of the exponential Golomb coded successive quantization weighting matrix.

  Deriving each component W (i) of the quantization weighting matrix using the decoded value d (i). At that time, processing is performed using the following mathematical formula. In the following equation, M is 16 for a 4 × 4 transform block and 64 for an 8 × 8 transform block.

W (0) = d (0) +8
If ((d (i) <128) && (d (i)>-128) && ((W (i-1) + d (i) +256)% 256! = 0))
・ W (i) = (W (i-1) + d (i) +256)% 256, for 0 <i <M
Otherwise ・ W (i) = W (k-1), for i = k,… <M, where D (k)> 128 or D (k) <-128, or W (k) = 0, and 0 <i, k <M
[W [I] cannot be 0]

  The quantization weighting matrix W (i, j) is determined from W (i) using an inverse block scan process. FIG. 5A shows the scan order in the 4 × 4 transform block, and FIG. 5B shows the scan order in the 8 × 8 transform block.

  Furthermore, the above "1. Inverse processing of cyclic difference encoding" can be realized as the following quantization weighting matrix derivation processing. Here, “ScalingList” indicates a quantization matrix WM, “lastScale” indicates W [i−1], and “delta_scaling” indicates D [i].

scaling_list (scalingList, sizeOfScalingList, useDefaultScalingMatrixFlag) [
lastScale = 8
nextScale = 8
for (j = 0; j <sizeOfScalingList; j ++) [
if (nextScale! = 0) [
delta_scale
nextScale = (lastScale + delta_scale + 256)% 256
useDefaultScalingMatrixFlag = (j = = 0 && nextScale = = 0)
]
scalingList [j] = (nextScale = = 0)? lastScale: nextScale
lastScale = scalingList [j]
]
]

  INDUSTRIAL APPLICABILITY The present invention is suitable for an encoding device that encodes or decodes an image, a decoding device, a web server that distributes a moving image, a network terminal that receives the moving image, a digital camera capable of recording and reproducing moving images, and a mobile phone with a camera Suitable for telephones, DVD recorders / players, PDAs, personal computers, etc.

FIG. 1 is an explanatory diagram showing an example of the data structure of a bitstream. FIG. 2 is a diagram illustrating an example of a quantization matrix. FIG. 3 is a block diagram illustrating a configuration of the image encoding device. FIG. 4 is a block diagram of the WM encoding unit. FIG. 5A is a diagram illustrating an encoding order of a quantization matrix. FIG. 5B is a diagram illustrating the coding order of the quantization matrix. FIG. 5C is a diagram illustrating an encoding order of a quantization matrix. FIG. 5D is a diagram illustrating an encoding order of a quantization matrix. FIG. 6A is a diagram illustrating a specific example of component values of a quantization matrix. FIG. 6B is a diagram illustrating a specific example of the difference value in the quantization matrix encoding process. FIG. 6C is a diagram illustrating a specific example of an encoding process from a frequency component value to a difference value and a variable length code. FIG. 7 is a flowchart showing the WM encoding process in the WM encoding unit. FIG. 8A is a diagram illustrating a specific example of variable length coding in the case of a positive number. FIG. 8B is a diagram illustrating a specific example of variable length coding in the case of being signed. FIG. 8C is a diagram illustrating a specific example of variable length coding. FIG. 9A is a diagram illustrating an arrangement of frequency components of a quantization matrix. FIG. 9B is a diagram showing how the encoded data of each component of the quantization matrix is arranged in the header part. FIG. 9C is a diagram showing how the encoded data of each component of the quantization matrix is arranged in the header part. FIG. 10 is a block diagram illustrating a configuration of the image decoding apparatus. FIG. 11 is a block diagram illustrating a configuration of the WM decoding unit. FIG. 12 is a flowchart showing the WM decoding process in the WM decoding unit. FIG. 13A is an explanatory diagram of a recording medium for storing a program for realizing the image encoding method and the image decoding method according to the above-described embodiment by a computer system. FIG. 13B is an explanatory diagram of a recording medium for storing a program that implements the image encoding method and the image decoding method according to the above embodiment using a computer system. FIG. 13C is an explanatory diagram of a recording medium for storing a program that implements the image encoding method and the image decoding method according to the above embodiment using a computer system. FIG. 14 is a block diagram showing the overall configuration of the content supply system. FIG. 15 is a diagram illustrating a specific example of a mobile phone using the image encoding method and the image decoding method. FIG. 16 is a block diagram of a mobile phone. FIG. 17 is a diagram illustrating an example of a digital broadcasting system.

Claims (26)

An image encoding method for encoding an image by orthogonal transform and quantization in units of blocks and encoding a quantization matrix for deriving a quantization step corresponding to each frequency of the orthogonal transform coefficient,
For each frequency component of the quantization matrix, obtain a difference value between the frequency component and a predetermined value corresponding to the frequency component,
Encoding the difference value into a variable length code;
The variable-length code has a code length that is shorter or has the same length as the difference value is smaller. The image encoding method according to claim 1, wherein the predetermined value is a value of a frequency component corresponding to a difference value obtained immediately before. The image encoding method according to claim 2, wherein the difference value is obtained in order from a low frequency to a high frequency of frequency components included in the quantization matrix. The image encoding method according to claim 3, wherein the difference value is obtained as a remainder divided by 2 to the power of k (k is a constant). The image encoding method further includes:
Determine whether successive difference values 0 exist at the end of the quantization matrix;
The image coding method according to claim 3, wherein when it is determined that there is a continuous difference value 0, the continuous difference value 0 is not variable-length encoded. The encoding method further includes:
6. The image encoding method according to claim 5, wherein an end code is added at the end of the variable length encoding, and the subsequent difference values are 0 which are not encoded. An image decoding method for decoding an encoded image in units of blocks by inverse quantization and inverse orthogonal transform, and decoding an encoded quantization matrix,
The variable-length-coded quantization matrix is variable-length decoded into a difference value for each frequency component,
The frequency component of the quantization matrix is obtained by adding the difference value and a predetermined value corresponding to the frequency component,
The variable length code has a code length that is shorter or has the same length as the difference value is smaller. The image decoding method according to claim 7, wherein the predetermined value is a value of a frequency component obtained by immediately preceding addition. The image decoding method according to claim 8, wherein the order of the addition is from a low frequency to a high frequency of frequency components. The image decoding method according to claim 9, wherein the frequency component is obtained as a remainder divided by 2 to the power of k (k is a constant). In the image decoding method,
When the end code is detected from the variable-length-coded quantization matrix, the same value as the frequency component immediately before the end code is output as the value of each subsequent frequency component. Item 10. The image decoding method according to Item 9. An image encoding apparatus for encoding an image by orthogonal transform and quantization in units of blocks, and encoding a quantization matrix for deriving a quantization step corresponding to each frequency of the orthogonal transform coefficient,
For each frequency component of the quantization matrix, a subtraction means for subtracting a difference value between the frequency component and a predetermined value corresponding to the frequency component;
Encoding means for encoding the difference value into a variable length code,
The variable-length code has a code length that is shorter or has the same length as the difference value is smaller. The image coding apparatus according to claim 12, wherein the predetermined value is a value of a frequency component corresponding to a difference value obtained immediately before. The image encoding device according to claim 13, wherein the difference value is obtained in order from a low frequency to a high frequency of frequency components included in the quantization matrix. The image encoding device according to claim 14, wherein the difference value is obtained as a remainder obtained by dividing by 2 to the power of k (k is a constant). The image encoding apparatus further includes a determination unit that determines whether or not a continuous difference value 0 exists at the end of the quantization matrix,
The image encoding device according to claim 14, wherein the encoding unit does not perform variable length encoding on the continuous difference value 0 when it is determined that there is a continuous difference value 0. The encoding unit further adds an end code to the end of the variable-length encoding, and indicates that the subsequent differential values are 0 that are not encoded. Image encoding device. An image decoding device that decodes an encoded image in units of blocks by inverse quantization and inverse orthogonal transform, and decodes an encoded quantization matrix,
Decoding means for variable length decoding the variable length encoded quantization matrix into a difference value for each frequency component;
Adding means for obtaining the frequency component of the quantization matrix by adding the difference value and a predetermined value according to the frequency component;
The variable length code has a code length that is shorter or the same as the difference value is smaller. The image decoding apparatus according to claim 18, wherein the predetermined value is a value of a frequency component obtained by the previous addition. The image decoding apparatus according to claim 19, wherein the order of the addition is from a low frequency to a high frequency of frequency components. The image decoding device according to claim 20, wherein the frequency component is obtained as a remainder obtained by dividing by 2 to the power of k (k is a constant). The image decoding device further includes:
Holding means for holding the frequency component obtained immediately before by the adding means;
Detecting means for detecting an end code from a variable-length-coded quantization matrix;
The image decoding according to claim 21, wherein when the end code is detected, the holding circuit outputs the value of the frequency component immediately before the end code as the value of each subsequent frequency component. Device. A program for causing a computer to execute an image encoding method for encoding an image by performing orthogonal transform and quantization in units of blocks, and encoding a quantization matrix for deriving a quantization step corresponding to each frequency of the orthogonal transform coefficient Because
The program is
For each frequency component of the quantization matrix, obtain a difference value between the frequency component and a predetermined value corresponding to the frequency component,
Encoding the difference value into a variable length code is executed by a computer,
The variable length code has a code length that is shorter or the same as the difference value is smaller. A program for causing a computer to execute an image decoding method for decoding an encoded image by inverse quantization and inverse orthogonal transform in units of blocks and decoding an encoded quantization matrix,
The program is
The variable-length-coded quantization matrix is variable-length decoded into a difference value for each frequency component,
By adding the difference value and a predetermined value according to the frequency component, the computer is caused to calculate the frequency component of the quantization matrix,
The variable length code has a code length that is shorter or the same as the difference value is smaller. A semiconductor device for encoding an image by orthogonal transform and quantization in units of blocks, and encoding a quantization matrix for deriving a quantization step corresponding to each frequency of the orthogonal transform coefficient,
For each frequency component of the quantization matrix, a subtraction means for subtracting a difference value between the frequency component and a predetermined value corresponding to the frequency component;
Encoding means for encoding the difference value into a variable length code,
The variable-length code has a code length that is shorter or the same as the difference value is smaller. A semiconductor device which decodes an encoded image by inverse quantization and inverse orthogonal transform in units of blocks, and decodes an encoded quantization matrix,
Decoding means for variable length decoding the variable length encoded quantization matrix into a difference value for each frequency component;
Adding means for obtaining the frequency component of the quantization matrix by adding the difference value and a predetermined value according to the frequency component;
The variable-length code has a code length that is shorter or the same as the difference value is smaller.

Images